Hydrogen embrittlement is a phenomenon in which mechanical properties of metallic materials, such as tensile strength and ductility, deteriorate due to the uptake of hydrogen. Such degradations decrease the fracture resistance of metals such as steel.
The hydrogen atom ranks as the smallest in diameter among the elements. Hydrogen atoms are easily adsorbed on metal surfaces from which they diffuse into the interior by jumping between the interstitial lattices of tetrahedral/octahedral sites. Hydrogen can be also trapped at metallurgical defects and imperfections in steel such as grain boundaries, dislocations, inclusions etc. Once atomic hydrogen is absorbed, it may precipitate at high-stress zones such as defects, inclusions, voids or discontinuities where a recombination reaction can take a place. The recombination can cause embrittlement, leading eventually to cracking. As hydrogen accumulates, linkage of such high-stress zones allows cracks to propagate through the metal.
Hydrogen diffusivity (DH) is a property of the metal that determines the rate at which hydrogen travels in the material and plays a major role in hydrogen damage development. Hydrogen damage such as hydrogen-induced cracking is likely to grow faster in high diffusivity materials due to an increased pressure build-up rate. Therefore, accurate knowledge of DH for a specific material of interest is a crucial input to hydrogen damage evolution models and lifetime prediction tools.
The review of the related art shows that there is a significant discrepancy between published values of DH even for the same steel grade (see table 1). For example, for X65 pipeline steel, reported values of diffusivities range from 10−5 cm2/s to 10−7 cm2/s, which is a variation of two orders of magnitude. The large discrepancies can be explained by multiple factors such as the difference in steel microstructure, specimen thickness, surface preparation and permeation test conditions.
TABLE 1STEELEXPERIMENTALHYDROGEN DIFFUSIVITYTYPEMETHODat 25 C. (cm2/s)X65Permeation9.49 × 10−7X52X70Permeation0.1-0.9 × 10−7   0.1-0.3 × 10−7   X100Permeation1.04 × 10−8X80Permeation5.32 × 10−9X70Permeation2.63 × 10−7X65Permeation 1.5 × 10−6X120Permeation2.0-2.8 × 10−7   X65Permeation0.8-2.7 × 10−9   X-52Permeation7 × 10−7 to 3.2 × 10−5X-60Permeation5.6-11.5 × 10−7   X-650.9-4.6 × 10−7   X-804.7 × 10−7X-1003.9 × 10−7X-654.2 × 10−7X-854.0 × 10−7X-60Permeation3.5 × 10−6
Due to this large variability in measured DH values, it is important that DH be measured directly on a portion of the metallic structure of interest to ensure accuracy. The directly measured value of DH can then be used as an input to a prediction model. The main challenge for such direct measurement is that the standard measurement technique, as described in ISO 17081, is destructive in nature, as it requires extracting and machining a test specimen from the equipment of interest. Obtaining a test specimen in this manner is usually impossible for installed and operational metallic structures.
The standard technique of ISO 17081 is based on the use of the electrochemical cell of Devanathan and Stachurski, shown in FIG. 1A. An electrochemical cell 100 includes a charging cell 110 and an oxidation cell 120. The charging cell 110 includes a platinum auxiliary (counter) electrode 114 and a calomel reference electrode 118. Similarly, the oxidation cell 120 includes a platinum auxiliary (counter) electrode 124 and a calomel reference electrode 128. A sample 130 is placed between the charging cell 110 and the oxidation cell 120. In operation, the charging cell 110 induces generation of hydrogen on the side of the sample surface 130 exposed to the charging cell. Some of the hydrogen generated on the charging cell side diffuses through the sample to the oxidation cell 120, where the hydrogen atoms are oxidized. The oxidation process is facilitated by keeping the sample 130 at a positive potential of around (+300 mV) against the standard calomel electrode 118. The use of palladium coating at the exit side can enhance the oxidation process further. The oxidized hydrogen is measured at an outlet port as a function of the oxidized current density. From the curve of oxidation current over time, DH is typically calculated using the time lag method.
The time lag method is appropriate for determining DH over a single dimension, e.g., the diffusivity of hydrogen from one side of a specimen to the other. It is derived from the one-dimensional diffusion equation. The analytical solution for the transient permeation flux is provided in ISO-17081 as:
                                                        J              perm                        ⁡                          (              t              )                                            J            SS                          =                  1          +                                    ∑                              n                =                1                            ∞                        ⁢                                                            (                                      -                    1                                    )                                n                            ⁢                              exp                ⁡                                  (                                                            -                                              n                        2                                                              ⁢                                          π                      2                                        ⁢                    τ                                    )                                                                                        (        1        )            where Jperm(t) is the transient permeation flux, JSS is the steady state permeation flux (i.e, JSS=Jperm(t=∞)) and τ is the normalized time expressed as function of the diffusion coefficient and the specimen thickness L as follows:
                    τ        =                              D            H                                tL            2                                              (        2        )            
A plot of the normalized permeation flux
  (            J      perm              J      SS        )versus normalized time τ is illustrated in FIG. 1B (with a log scale on the time axis). The usefulness of this curve (hereafter referred to as the ‘standard master curve’) is that it is independent of the hydrogen charging concentration C0, the specimen thickness L, and the hydrogen diffusivity DH. In other words, for any hydrogen permeation experiment that satisfies the one-dimensional conditions shown in FIG. 1A, a plot of
  (            J      perm              J      SS        )vs. τ will stick to this standard master curve. Tabulated values of this curve are provided in the ISO standard 17081, such that if the specimen thickness L and the permeation transient Jperm(t) are known, the hydrogen diffusivity DH can be easily determined from any point on the curve. In practice, the point on the curve where
            J      perm              J      ∞        =  0.63is commonly used. This point corresponds to a normalized time
  τ  =            τ      lag        =                  1        6            .      The physical time corresponding to τlag is noted tlag and is therefore by definition equal to
      t    lag    =                    6        ⁢                  L          2                    DH        .  From me latter me hydrogen diffusivity DH is easily derived as:
                    DH        =                              L            2                                6            ⁢                                                  ⁢                          t              lag                                                          (        3        )            
The standard technique discussed above requires a test specimen and access to both sides of the specimen. As noted, this technique is not applicable to determining hydrogen diffusivity of metal structures in the field or to multi-dimensional hydrogen permeation flux. There is therefore a need for a non-destructive measurement technique able to carry out on-site and in-service measurement of hydrogen diffusivity as required. The embodiments of the present invention address this need.